INDEX

Explanations

terms related to changes and reforms in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

changes and updates

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

WriteTagHelper

-0.47

IUrlHelper

-0.43

 propOrder

-0.40

Hentet

-0.39

Compliance

-0.39

kose

-0.38

 Lend

-0.37

colhead

-0.37

 تضيفلها

-0.37

quiry

-0.36

POSITIVE LOGITS

tonode

0.53

 يتيمه

0.47

şört

0.45

 autorytatywna

0.45

 <<<<<<<<<<<<<<

0.44

EDEFAULT

0.44

 Menschheit

0.42

Diweddarwch

0.41

 Betracht

0.39

 cordones

0.38

Activations Density 0.051%

terms related to changes and reforms in various contexts

changes and updates

No Comments

No Known Activations